Seasonal Affective Disorder and associated Factors among Sri Lankan Immigrants living in Oslo

(1)

SEASONAL AFFECTIVE DISORDER AND ASSOCIATED FACTORS AMONG SRI LANKAN IMMIGRANTS LIVING IN

OSLO

T.B.Saheer

Supervisor: Bernadette N. Kumar

Co-supervisors: Edvard Hauff, Lars Lien and Karin Harsløf Hjelde

Faculty of Medicine

Institute Of General Practice and Community Medicine Section for International Health

May 2012

Thesis submitted as a part of the Master of Philosophy Degree in International

Community Health

(2)

Acknowledgements

My heartfelt gratitude should go, first of all to Dr.Bernadette Kumar, my supervisor. She was behind me throughout this project, spent numerous hours with me to fulfill my task and her encouragement and guidance was invaluable to me.

My co-supervisor Karin Harsløf Hjelde guided me through the qualitative part of the study and her efforts and encouragement should be warmly appreciated. Prof.Edvard Hauff and Dr.Lars Lien spent their invaluable time with me and their comments and support helped me a lot and I was lucky to have them as my co-supervisors.

Everybody at NAKMI helped me in numerous ways to carry out this project successfully and a big thank you for them.

My deepest gratitude go to all the participants who spent their time and shared their experiences with me in the qualitative part of the study, they made this work a reality. Ajith and Sabine, my friends should be remembered with gratitude for their help and time spent with me for the project.

The MPhil coordinators and all the staff and colleagues at the Institute of Public Health at the University of Oslo are warmly remembered for their help and support.

I am also indebted to my family and friends who took care of me during these two years in Oslo.

(3)

Abstract

Human migration is as old as the human evolution. Factors influencing migration are numerous and the number of migrants is increasing worldwide. Immigrants constitute 13.1% of the Norwegian population and 23% of the population of the capital of Norway, Oslo. Migration can be associated with increased risk of mental illness.

Seasonal affective disorder (SAD) is more prevalent in populations who live in higher latitudes. SAD is described to occur more in the winter season (W-SAD) than in the summer. In addition a milder form of SAD, Sub-syndromal SAD (S-SAD) is also described in the literature. Despite the fact that SAD among immigrants found to be higher than the native population in some studies, there is a significant paucity of research on SAD among immigrants.

This is the first study on SAD among the immigrants in Norway. The quantitative part of the study used the data collected among adult (31 to 60 years) immigrants living in Oslo in 2002 (The Oslo immigrant health study). The qualitative part of this study conducted in 2011, which includes eight in-depth interviews among adult Sri Lankan immigrants living in Oslo, is novel as there are very few qualitative studies to look in to seasonal affective disorder.

There were significant differences in SAD prevalence rates among the five immigrant groups (Turkey;

16.9, Sri Lanka; 6.9, Iran; 21.5, Pakistan; 17.7 and Vietnam; 14.9) included in the study. W-SAD was significantly associated with country of birth, younger age, smoking, presence of mental distress, frequent visits to the psychiatrist or the General Practitioner (GP), self reported poor health and presence of chronic illnesses. Gender, Number of years living in Norway, education and employment status were not significantly associated with W-SAD. S-SAD was significantly associated with Country of birth, smoking and alcohol consumption.

Sri Lankan immigrants expressed the view that seasonal changes do not affect their mood and they feel happy and contented about their physical and mental health. They also described that family, social and cultural integrity and better economic prospects as reasons behind their perceptions of happiness.

In conclusion, Iranians had the highest and the Sri Lankans had the lowest prevalence of W-SAD, and prevalence of SAD is not as high as compared to other studies among immigrants. These findings were confirmed by the qualitative study, where Sri Lankan immigrants attributed lower levels of SAD to close family and social networks and better economic prospects. Further research on perception of SAD and mental health among immigrants, especially the other four groups, should be encouraged.

(4)

List of Abbreviations

BMI Body mass index

FGD Focus group discussion

GOSL Government of Sri Lanka

GP General Practitioner

GSS Global seasonality score HSCL Hopkins symptom check list

HUBRO Oslo Health Study

Innvandrer HUBRO Oslo immigrant health study LDL cholesterol Low density lipoprotein cholesterol MDGs Millennium development goals

PO Participatory observation

REK Regional Committee for Medicine and Health Research Ethics, Oslo S- SAD Sub syndromal Seasonal affective disorder

SAD Seasonal affective disorder

SPAQ Seasonal pattern assessment questionnaire

SSI Seasonality score index

Summer-SAD Summer- Seasonal affective disorder

UiO University of Oslo

UN United Nations

WHO World health organization

W-SAD Winter-Seasonal affective disorder

(7)

List of tables

Table 1: Total population for country ... 37

Table 2: Number and percentages of different variables for the total sample ... 38

Table 3: Mean value and Standard Deviation (SD) of age, duration of living in Norway, Average HSCL score, and Total score GSS for the total sample ... 39

Table 4: Mean value and Standard Deviation (SD) of age, duration of living in Norway, Average HSCL score, and Total score GSS for different groups ... 39

Table 5: Prevalence rates of W-SAD and S-SAD for different groups... 40

Table 6: Results of the Chi squared test with independent variables and S-SAD ... 40

Table 7: Results of the chi squared test with independent variables and W-SAD ... 41

Table 8: Results of the Logistic Regression analysis for W-SAD ... 42

Table 9: Results of the Logistic Regression analysis for S-SAD ... 43

Table 10: Mean values of age, Number of years lived in Norway, Total GSS, Average HSCL score and Percentage of females in the Total sample (n=3019), Analyzed sample (n=1047) and the removed sample (n=1972) ... 43

Table 1: Total population for country ... 81

Table 2: Number and percentages of different variables for the total sample ... 82

Table 3: Mean value and Standard Deviation (SD) of age, duration of living in Norway, Average HSCL score, and Total score GSS for the total sample ... 83

Table 4: Mean value and Standard Deviation (SD) of age, duration of living in Norway, Average HSCL score, and Total score GSS for different groups ... 83

Table 5: Prevalence rates of W-SAD and S-SAD for different groups... 83

Table 6: Results of the Chi squared test with independent variables and S-SAD ... 84

Table 7: Results of the chi squared test with independent variables and W-SAD ... 84

Table 8: Results of the Logistic Regression analysis for W-SAD ... 85

(8)

Table 9: Results of the Logistic Regression analysis for S-SAD ... 86 Table 10: Mean values of age, Number of years lived in Norway, Total GSS, Average HSCL score and Percentage of females in the Total sample (n=3019), Analyzed sample (n=1047) and the removed sample (n=1972) ... 86

(9)

1. INTRODUCTION

1.1 Migration

The phenomenon of migration, simply defined as seasonal movement of animals from one place to another is observed in smaller creatures like frogs to the world’s largest living thing, the blue whales (1, 2). In contrast to animals whose migration is seasonal, migration of the modern man varies from escape from torture to education or finding better employment (3).

History of human migration is as interesting and fascinating as the human evolution. The journey of human migration started approximately 100 000 years ago at the birth place of human kind, the African continent causing humans to spread far and wide all over the globe, and this remarkable phenomenon still exists among human beings though in an entirely different manner (4,5). In contrast to the primitive man who principally migrated to find more food or to avoid harsh weathers, migration of the modern man is far more complicated and diverse. Modern migration, which can be defined as the permanent or long term change of residence by an individual or a group has different reasons (6). These can be economical reasons such as employment, educational needs or political reasons, ranging from escape from torture and persecution to avoiding discrimination such as different religious beliefs (3, 6, 7).

Type of migration varies from one context to another and migration could occur within the borders of the same country or people could migrate to other countries crossing international borders (3, 4, 6).

International migration has grown rapidly during the last few decades. In 1965 there were 75 million migrants in the world which increased to 84 million during the next decade and by the year 2007 the number increased to 175 million people (4). According to the UN (United Nations) the total number of international migrants in 2009 was 214 million (about 3% of the world population) and 15 million of them were refugees. This trend is expected to be continued and may be even escalated due to the recent upheavals of civil wars in many parts of the world. It is estimated that by the middle of the 21^st century the total number of international migrants will be around 230 million (6, 8, 9).

1.2 Migration to Norway

The very first group of Norwegians migrated to the North America in the year of 1825. Since then Norway used to be the sending country of migrants rather than the opposite during the nineteenth century and the early twentieth century (10). Nearly 800 000 Norwegians or almost

(10)

half of the population of the country (population of Norway in 1865 was 1.7 million) emigrated during this period (10, 11). But with the boom in the oil industry and subsequent economic development Norway attracted, first, more labor migrants and later refugees and asylum seekers with the escalation of conflicts in Asia, Latin America and Africa (10). Now, according to the Statistics Norway, more than half a million people with an immigrant background live in Norway (655000; 13.1 % of the total population) and they have migrated from 215 different countries and independent regions of the world. Between 1990 and 2010, 471 000 non-Nordic citizens were granted residence in Norway and 22% of them were refugees and another 28% were labor migrants (11% migrated on education purposes and 38% on family reunification) (11, 12).

Sri Lanka is a small island in the Indian Ocean with a population of 20 million. Three major ethnic groups live in the Island; Sinhalese, Tamils and Muslims (13, 14). The Island was ravaged by a civil war that began in the early 1980s between the Government of Sri Lanka (GOSL) and the Tamil rebel fighters which lasted nearly 30 years displacing tens and thousands of civilians of all the three ethnicities. Majority of the displaced civilians were Tamils and they migrated all around the world mainly as asylum seekers or refugees.

A small group of Sri Lankan Tamils have migrated to Norway in the 1960s as labor migrants for the fishing industry and had gained the reputation of hard working people (15). But the majority of Sri Lankans migrated after the escalation of the war in the 1980s. At the beginning, most of the initial Tamil refugees settled down initially in the Finmark County because there were more job opportunities in the fishing industry in the Northern Norway.

They were recognized as hard working and had better wages than people living in the south and their living conditions were similar to those of the native Norwegians. Soon they built their own little community with separate movie theatre, radio station, shops and cafes (10).

According to the Statistics Norway, by the 1^st of January 2012, 14293 Sri Lankan immigrants were living in Norway (first generation immigrants and Norwegian born to Sri Lankan immigrant parents). They rank 16^th on the immigrants’ list and more than half of this population (7365) is living in Oslo (16). Among the immigrant population of Norway, compared to many of the other immigrant groups, Sri Lankan immigrants are reported to be employed more in the labor market and also enjoy relatively higher income levels, lower crime rates, and higher education levels (11, 15).

(11)

1.3 Migration and mental health

The ancient Latin proverb of the Greek philosopher Thales “mens sana in corpore sano” (a sound mind in a sound body) is well reflected in the WHO definition of “health”, in which mental health is regarded as one of the essential components of overall health of an individual. Mental health of an individual is in turn related to and affects socio-cultural, economical and physical well being. On the other hand mental health of an individual is determined by socio- economical, cultural, environmental, biological and genetic factors. A person with a good mental health should be able to work productively according to his or her own abilities, cope with normal stressful events in life and contribute meaningfully to the community in which he or she lives (6, 17, 18).

According to the WHO statistics more than 450 million people worldwide suffer from mental disorders and it is estimated that many more suffer from undiagnosed mental illnesses. In the year 2008, reported number of deaths due to neuropsychiatric conditions were more than 1.3 million and it was 2.3% of the total number of deaths worldwide (19). Poor mental health is associated with multiple factors such as major life style changes (retirement from work), socio cultural factors (war and conflicts), personal factors (bereavement of a close relation or a friend), political reasons (persecution or migration) and environmental factors (natural disasters) (6). Mental illnesses, while directly contributing to the global burden of disease, indirectly affect the health of the population in multiple ways. The importance of mental health on global burden of disease becomes obvious when we analyze the Millennium Development Goals (MDGs). Mental health, though not included in the MDGs has been shown to be very important for the successful implementation of the MDGs (20, 21). In addition mental health is also associated with suicides, accidents and injuries and increased risk of chronic diseases such as diabetes, hypertension and myocardial infarction (22, 23). It is also associated with reduced immunity and increased risk of infections (6). The importance of mental health was recognized by the WHO by declaring the year 2001 as “the global mental health year”.

Migration has many benefits as well as many disadvantages for the migrant, the host country and the home country (6). Safety, improved socio-economic conditions and better educational opportunities are some of the benefits a migrant can experience in the host country. On the other hand, loss of own cultural and religious support, loss of autonomy and increased risk of illnesses can put the migrants in a disadvantageous situation in the host country (3, 6).

(12)

Migration is described as a three stage process; pre-migration, migration and post-migration (3, 6, 24). Different factors affect an individual’s health during these three different stages of the migratory process. Different pull factors (such as better socio-economic conditions in the host country) and push factors (such as violence and persecution in the home country) influence the decision of migration (3, 24). Migration also can be categorized as voluntary or involuntary depending on the context of migration. Labor migration, most of the time is undertaken by the migrant voluntarily in contrast to forced migration faced by asylum seekers and refugees and the health impact of migration on the migrant vary enormously in these two different groups (3, 6, 7).

Health impact especially the mental health impact of migration on the individual and community has been studied for decades. Migration is associated with a higher risk of mental and physical illnesses among immigrants compared to the host population irrespective of reason for migration (3, 6, 18). In contrast to earlier beliefs, even the individuals who have migrated voluntarily to a new place such as labor migrants are increasingly recognized as having higher risk of illnesses than the host population despite the fact that they enjoy relatively much higher standards of living compared to the individuals who migrated as refugees and asylum seekers from the same community (6, 25).

Migration can cause an enormous psychological stress on the migrant and the families (3, 26).

During the pre-migratory stage violence and persecution in the home country and traumatic experiences can increase the risk of mental illness in an individual. Hazardous border crossings, anxious waiting and even imprisonment of a migrant are some of the factors that can impact seriously on an individual’s mental health during the second stage of the migratory process. Once a migrant has arrived to the host country number of different factors can adversely affect the mental health of a migrant. Initially migrants may have a sense of relief especially those who escaped conflicts, violence and torture. But gradually other issues would emerge causing much anxiety and distress to the individual. Immigrants are living between two cultures or sometimes even more than two cultures (6). Their family and social structure may be destroyed and they live in a foreign land with a different culture, society and language. All these factors create psychological stress on the immigrants and their families which would lead to mental ill health (3, 6, 24). However an individual’s reaction to the stress of migration depends on several factors such as personality, psychological stability and cultural identity of the individual. On the other hand immigrants are a vastly heterogeneous

(13)

group and vary enormously in handling the stressors of migration and not all immigrants have the same experiences and they are not affected to the same degree of mental ill-health (26).

Social support and acceptance by the host society and the own ethnic group and better employment and economic status in the host country also play a major role in shaping an immigrant’s mental health (3, 6, 26).

The “healthy migrant effect” described in earlier literature as immigrants having better health status than the native population has now being increasingly questioned (27). Some first generation immigrants in fact had been found to be healthier than the host population when they first arrived to the new country. But the health status of the immigrants deteriorates with time due to several factors such as poor living conditions, unemployment, adoption of an unhealthy life style and poor health care provision and utilization (27). On the other hand it is described that immigrants are reluctant to use some of the health care facilities available in the host country especially mental health care facilities (28). One reason could be that the amount of stigma associated with mental health in the immigrants’ native culture (15). In the Sri Lankan society the attitude towards mental health patients found to be negative even among health personal (29).

1.4 Seasonal Affective Disorder (SAD)

Changes in mood and behavior in different seasons is called seasonality (30, 31). Seasonality of disease occurrence has been described since ancient times (32). Seasonality is found to be a universal phenomenon affecting all individuals to a certain degree and the Seasonal Affective Disorder (SAD) which was first described as a separate clinical entity in 1984 is at the extreme end of seasonality (33). SAD could occur in the summer (Summer-SAD) and winter (W-SAD) though the latter has been studied most often. W-SAD is characterized by typical depressive mood and atypical symptoms such as increased sleep, increased appetite, increased weight and carbohydrate craving (33, 34). SAD has been considered as a specifier of bipolar or recurrent major depressive disorder in the DSM 4 criteria (35). Major risk factors for SAD are living in higher latitudes, amount of sunlight per day, female gender and younger age.

Socio-cultural factors, climate, ethnicity and genetic factors have also been associated with SAD and living with a partner, older age and male gender are described as protective factors.

Higher education level and higher income level of an individual has also been found as positively associated with higher levels of SAD (36-38).

(14)

Melatonin secretion in the brain which is associated with diurnal rhythm and neurotransmitters such as serotonin, norepinephrin and dopamine are suggested as biological mechanisms for the onset of SAD (30, 39). Prevalence of SAD varies from one geographical region to another even in the same country (34). Prevalence of W-SAD has been found to be higher in temperate countries in North America and Europe and the opposite is true for Summer-SAD which is comparatively higher in warmer climates (40-42). On the other hand prevalence rates of Sub-syndromal SAD (S-SAD) which is a milder form of SAD has been found to be generally higher than W-SAD prevalence (34). Exposure to bright light is the best and the most effective treatment of W-SAD (33, 34). Nonetheless, W-SAD is often under- diagnosed or miss-diagnosed and treated with expensive drugs despite the availability of this low cost and non-pharmacological method of treatment (43, 44).

Studies among immigrants on SAD are lacking and the available data shows that the prevalence of this disorder is higher among immigrants than the native population (45, 46). It is also higher among individuals who move to higher latitudes in the same country or region despite their ethnicity (34). Most of these studies had been conducted, however among small sub groups of immigrants such as students who have not lived in the new locations for a considerable time period (45, 46). So the results obtained in many of these studies could not be generalized.

Norway with its harsh seasonal changes and high immigrant population could be a typical place to experience SAD. Prevalence of W-SAD among the Norwegian native population varies between 6.5% to 19% in different studies (34, 47, 48). S-SAD prevalence among native Norwegian population found to be 10.1% among males and 10.8% among females (34). I could not find any published studies on SAD among any immigrant groups living in Norway.

I also could not find any published qualitative study on SAD among immigrants among the available literature neither in Norway nor in other countries.

1.5 Need for more research among immigrant groups

Immigrants’ health conditions and health needs are increasingly recognized as being different than the host population (6, 24, 49). Immigrants come from different social and cultural backgrounds with different health beliefs, attitudes and needs. In addition most of the studies that had been carried out among immigrants have considered immigrants as a single homogenous group. This has led to making blanket health policies covering all immigrants in

(15)

the disease prevalence rates vary considerably among different immigrant groups (6, 7, 50).

On the other hand most of the studies have used instruments which are not culturally validated for the immigrant group under investigation which has led to the doubt of validity of the outcome of most of these studies. So the need for more research among different immigrant groups is an essential and urgent need for health policy planning and public health preventive strategies (24, 28).

Research material on migrants’ mental health is scarce (28). In Norway public health research on immigrants’ mental health has only a short history. But the available literature shows that the immigrants, especially the non western immigrants have more mental health problems than the host population (50, 51). It is also shown that economic conditions and social support are important factors in mental health of immigrants (51). The underutilization of available health facilities, especially the mental health facilities by immigrants has also brought in to light in some of the literature (15, 28).

Though the statistics show that the Sri Lankan immigrants living in Norway have better social, economical and educational status, the health indices do not show the same amount of positivity (11, 50, 52). The Oslo Health Study (HUBRO) and The Oslo Immigrant Health Study (Innvandrer HUBRO) show that Sri Lankans have higher levels of risk factors for cardiovascular diseases such as increased LDL cholesterol levels and obesity (50). It has also shown that compared to the host population Sri Lankan immigrants have poor dietary habits and lesser amount of physical exercise (50). But on the other hand mental health of Sri Lankan immigrants found to be better than some of the other immigrant groups living in Norway (15, 50). Nevertheless in a study among Sri Lankan immigrants living in Northern Norway, Grønseth (2010) describes that Sri Lankan immigrants have many health related problems especially psychological problems for which they do not get proper attention from their doctors (53). In her opinion the gap between the immigrants’ attitude towards their illnesses and the Norwegian physicians who treat them has created a conflict which has resulted in further aggravating the situation of the immigrants, especially their psycho-social wellbeing. But on the other hand her focus on the study was the immigrants’ use of medical services rather than mental health per se (53).

In the light of these research evidence one could postulate that answers for some key questions remained unanswered.

(16)

 What is the explanation for the important finding that Sri Lankan immigrants have lower levels of mental distress compared to some of the other immigrant groups living in Norway?

 Is there any cultural, social or other factor which has a protective effect against mental illnesses among the Sri Lankan immigrants living in Norway?

 How do Sri Lankan immigrants perceive the seasonal variations especially the cold winter months in Norway and how do they cope up with it?

To find answers for some of these questions I proposed to carry out a qualitative study and to analyze a group of already collected quantitative data. The quantitative part of the study used already available data from the Oslo Immigrant Health Study (2002). The second part was a qualitative study among Sri Lankan immigrants living in Oslo.

(17)

2 Main Research Objectives

The research had two main objectives because it had two components; a quantitative and a qualitative part.

1. Assess the prevalence of Seasonal Affective Disorder (SAD) among immigrants living in Oslo, Norway.

2. To gain an understanding of the factors affecting the well being and mental health of Sri Lankan immigrants living in Oslo, Norway.

2.1 Specific Objectives

1.1 Assess and compare the prevalence rates of Seasonal Affective Disorder among five ethnic immigrant groups in Oslo (Sri Lanka, Pakistan, Turkey, Iran and Vietnam).

1.2 Determine the associated factors affecting the prevalence of SAD in these groups.

2.1 Understand the migrant Sri Lankans’ perceptions about their well being and mental health.

2.2 Understand the reasons behind these perceptions.

2.3 Understand the informants’ perception on any seasonal changes to their wellbeing and mental health.

2.4 Understand the reasons behind the informants’ perception of the seasonal changes to their wellbeing and mental health.

(18)

3 Methodology

Generally the main reason a research is carried out is to discover, understand and communicate the truth of the object being researched and this object under study could be a person, a disease, a phenomenon or a situation (54). The phenomenon called “truth” depends on the point of view of the researcher, the researched as well as the audience who view the results of the research. The word “research” also means differently to different individuals depending on his or her attitude towards research. The main two paradigms of research

“positivist” and “interpretivist” which use quantitative and qualitative research methods respectively had not been the same in the history of search for the “truth”. Before medieval times among the Greek and Roman philosophers the qualitative view dominated in the analysis and description of situations. During the middle ages a new way of reasoning began and only the “intellectually superior scholars” were thought to be able to fully comprehend and discuss the precise, meticulous and logical reasoning of a given subject. The common man and the non-intellectuals were regarded as not fully able to comprehend these and were not allowed to venture in to these fields. Frustrated with the situation where only the philosophers could express their reasoning, the scientist in the 17^th century came up with the idea of truth being self evident and observable. They believed that there cannot be any doubt about truth which could be quantitatively measured (54). So the debate about understanding and evaluation of truth raged on for centuries and finally the quantitative research methodology took shape in to a more course and effect relationship with a scientific background and qualitative research methodology was applied more in social sciences. So the gap between them widened and the interpretivist paradigm was not regarded as a true scientific method to apply on scientific research, but during the last few decades of the 20^th century qualitative research has again come to the attention proving its validity and usefulness in other branches of science such as public health (54-57).

The qualitative and quantitative debate during the last century can be roughly divided in to three stages (53). During the latter part of 19^th century up to the 1970s there was a stage of conflict and polarization between the two approaches. Quantitative and qualitative methods were regarded as having different theoretical positions and therefore used separate and specific methodologies of their own. They were considered to be so different that any attempt of bringing them together was looked in contempt. Especially in such fields like science and

(19)

medicine where precise, controlled and objective measurements were needed to obtain mathematically sound models to describe and illustrate the truth behind a given situation (54).

The next stage reached with a compromise between the two and both regarded as parallel research tools. The critical approach with good and bad viewpoint was relaxed and the differences were accepted as variations of the same concept. During the 1980s a more pragmatic stage was reached with cooperation of the two methods. Qualitative research was applied with more rigorous methodology so as to be regarded as a legitimate tool in scientific research. During the last two decades however the debate has died down a little and now both methods are regarded as complementary to each other and used in triangulation in research strengthening the possibility of reaching more accurate conclusions (54-56, 58, 59).

3.1 Quantitative research methodology

Quantitative research is used more in natural and applied sciences than the social sciences. It is more critical about the way a research is carried out and dictates stringent terms on how to apply scientific and logical approaches in research and how to analyze the data with mathematical and statistical tools. Quantitative research is more concerned with course and effect relationships of the situation under study and the goal of quantitative methodology is to measure the phenomena under study more objectively (54, 59). Much of the data we use in medicine such as demographics, prevalence rates and disease patterns are obtained from studies and surveys conducted under quantitative methods (56, 60). Almost all the information that is used in both communicable and non communicable diseases such as symptoms and signs, pathophysiology, diagnosis and treatment have been discovered based on quantitative methods. In quantitative research the researcher tries to minimize his or her influence on the research by using various scientifically accepted norms such as random sampling, blinding and controlling the confounders as much as possible (54, 60).

The positivist paradigm dictates that rationality belongs only to scientific knowledge and the methodology must be based purely on observations and any interference with interests, values or purposes are not tolerated (56, 57). Due to these strict criteria quantitative approach is criticized as mechanical and not having the human touch (55-57). The researcher and the researched or the interviewer and the interviewee are handled as objectively as possible to reduce their subjective influence on the outcome of the research. The research questions are made by the researcher before the research was carried out and there are stringent regulations

(20)

of both parties are suppressed so as to minimize any bias creeping in to the research study (57).

Two broad study design types are used in quantitative research of medical science;

experimental studies and observational studies (61). Experimental studies can generate the most convincing evidence in a research because the confounding factors which could affect the outcome of the research could be controlled to a much larger extent with this type of a study design. Observational studies, on the other hand delivers less reliable results because the confounding variables are not effectively controlled as opposed to experimental studies (61).

Public health research uses mostly observational studies. Several types of observational study designs are used in epidemiological surveys; cohort studies, case control studies and cross sectional studies (62). The cohort studies are longitudinal in nature and the defined sample population is studied over a longer time period either prospectively (forward in the time span) or retrospectively (backwards in the time span). In cohort studies disease patterns and cause and effect of a disease can be discovered. But these studies require more resources and time.

Case control studies are designed as retrospective studies. Cross sectional studies are described as snap shots of a single point in time (61, 62).

Cross sectional studies in contrast to longitudinal studies are easier to design, quicker to perform, and need fewer resources. Because these surveys are carried out in one point of time they can measure only disease prevalence rather than incidence of a disease which needed to be measured during a given period of time. Disease trends over time are also not measured due to the same reason. Cross sectional surveys can predict on risk associations of a disease but not any causations (61, 62). One way to overcome these limitations is to perform repeated cross sectional studies over a time period on the same study population. However it is not practical to get the same individuals as sample population and so the validity will be lost. On the other hand longitudinal studies need more time and resources and also could be hampered by recall bias of the subjects (61).

Innvandrer HUBRO study (The Oslo immigrant health study 2002)

I used the data collected in the Oslo Immigrant Health Study which was a large cross sectional survey. This survey included a target population of more than 7000 people and had multiple research objectives. For this type of a large study a cross sectional design would be more appropriate for several reasons. The large target population in the survey would have limited the options of the researchers due to the limitations in resources and time available

(21)

and if it was designed as a longitudinal study, follow up of the number of subjects participated would have been a daunting task both financially and logistically. On the other hand this study was part of a series of several studies and that also could have limited the options available for the research team. However this approach limits some of the epidemiological calculations and interpretations from the study outcome as discussed above.

The Oslo Immigrants Health Study (Innvandrer HUBRO) is a large cross sectional epidemiological survey and wasconducted in 2002 by the National Health Screening Service (now Norwegian Institute of Public Health – NIPH) and the University of Oslo (UiO). The target study population included all adult immigrant residents of Oslo, who were born in Turkey, Sri Lanka, Iran, Pakistan and Vietnam between 1942 and 1971 except the birth cohorts who were invited to another study previously (1954/55, 1960 and 1969/70). A total population of 7890 met the eligibility criteria. Out of this population 7607 were reached by mail to participate in the study. 3019 gave written consent and participated in the study. The response rate for the total population was 39.7% and the response rates for individual countries of birth were; Turkey 32.7 %, Sri Lanka 50.9%, Iran 38.8 %, Pakistan 31.7% and Vietnam 39.5%. Non responders were sent one written reminder between 3 to 8 months after the first invitation letter. Ethnicity, age and gender were determined by using the Norwegian population register and only the first generation immigrants were included in the survey.

The research group adopted several strategies to increase the participatory rate. The main questionnaire, consent form and the information sheet were translated in to the five respective languages of the participants, field workers at the screening centers spoke the five different languages, a mobile screening unit was used for the late responders, the project coordinator worked closely with immigrant groups, lectures and meetings were organized and radio and TV advertisements were broadcasted.

Objective of the main survey was to assess the main health problems of the adults of five of the largest immigrant groups living in Oslo and to compare their health status to the native Norwegian population.

Mass media was used to disseminate the information about the study and invitation letters were mailed to the eligible participants two weeks prior to the data collection. One main questionnaire and an additional supplementary questionnaire were completed by the participants and participants who had at least returned one questionnaire was included in the study. In addition to the two questionnaires, anthropometric measurements and blood tests

(22)

were performed on the subjects at several screening sites (but those data were not used in my analysis except the calculated body mass index (BMI)).

The main questionnaire included questions on physical and mental health, social activities, education, employment, alcohol consumption and smoking habits. The questionnaire was in Norwegian and also included a translated version of the native language of the participant.

The supplementary questionnaire (which was only in Norwegian and English) contained the Seasonal Pattern Assessment Questionnaire (SPAQ) which is used to calculate the prevalence of SAD.

3.1.1 Instruments

Seasonal pattern assessment questionnaire (SPAQ) which was included in the supplementary questionnaire is the commonest instrument used in epidemiological surveys of SAD (34, 63, 64). It has several scales and three of them were used in my analysis. Seasonality score index (SSI) is one of the mostly used scales. SSI investigates seasonal variation of six items (sleep, social activity, mood, weight, appetite and energy). These items are measured in a scale of 0 to 4 ranging from ‘no change’ to ‘major change’. The total score of these six items which ranges from 0 to 24 is called the global seasonality score (GSS). The second scale used asked the participants to rate the degree to which they experience the seasonal changes as a problem (ranging from ‘no problem’ to ‘completely disabling problem’). The subjects were also asked to rate which month of the year they felt the seasonal changes to be worst (31, 34, 38).

Total GSS equal to or more than eleven with the degree of problem as equal to or more than moderate degree is regarded as a SAD positive case. With SAD positive and if the subject felt worse in the winter months (November to February) then the subject is categorized as W- SAD. The same procedure is carried out for the Summer-SAD and a subject is categorized as Summer-SAD if the subject felt the seasonal changes worst in the summer months (May to August) (31, 34, 38).

Calculation of S-SAD uses two criteria. If the total GSS is more than 11 with the seasonal changes being a problem less than moderate degree then the subject is classified as S-SAD. If the total GSS is 9 to 10 and the subject feels the seasonal variation to be an equal to or more than a moderate problem then also the subject is classified as S-SAD positive (34). A subject who falls under any one of these two criteria is classified as S-SAD.

(23)

The Hopkins symptom check list- 10 item version (HSCL-10) is a widely used scale to assess the mental distress in epidemiological surveys (65, 66). This scale was included in the main questionnaire form. It uses a scale with responses ranging from ‘not troubled (1)’ to ‘much troubled (4)’ and assesses the experience of the subject on ten different items experienced during the past one week. The average of the total sum of the all ten items is used to classify mental distress. Sample mean of any missing item replaces the missing values. Nevertheless responses with more than 3 missing items were excluded from my analysis. Those scoring an average of 1.85 or more in the HSCL-10 are classified as mentally distressed (65, 66).

3.1.2 Ethical considerations

The approval and ethical clearance for the quantitative (Innvandrer HUBRO) study was obtained from the Norwegian Data Inspectorate and the Regional Committee for Medical Research Ethics (REK). The study has been conducted according to the ethical principles declared by the World Medical Association Declaration of Helsinki. All the participants gave written, signed consent for the study prior to the data collection. All the data were encrypted to ensure confidentiality of the participants.

3.2 Qualitative research methodology

Malterud (2001) defines qualitative research method as “systematic collection, organization and interpretation of textual material collected by talk or observation” (55). The basic aim in qualitative research is to obtain a comprehensive understanding of the situation under study (54). To gain this understanding a researcher can use different approaches in qualitative research. Four broad theoretical approaches are described in qualitative research; critical theory, feminist theory, postmodernist theory and interpretive constructionist theory (67).

In critical theory the researcher emphasizes action research, tries to highlight and uncover problems in the society such as problems of the oppressed poor people, migration and stigmatized diseases like HIV/AIDS. The researcher often takes the view point of the oppressed and argues that the knowledge is subjective and depends on the perceiver’s viewpoint (54, 67).

Empowering the disempowered groups like women are important for a feminist researcher.

They argue that the positivist paradigm with surveys as research methods disempower subjects and disregard the feelings and cultural influence of the subject under study. The

(24)

research methodology used by the feminist researchers is believed to open up interviewees to talk more freely and express themselves more coherently (54, 67).

For a postmodernist researcher truth is impossible to define and neutrality is also impossible to achieve because everyone involved in the research influences the whole process of research from writing the proposal to the publication of findings. Because different people see the same problem differently they accept different viewpoints from different researchers about the same situation under study. Many argue that only the viewpoint and the voice of the interviewee should be the outcome of a study (54, 67).

Interpretive constructionist theory states that interviewee’s views on experiences and observations about his or her life, work and society is more important. The researcher is more concerned with different views of different interviewees and tries to draw specific and relevant meanings of the interviews and interpret them comprehensively (54).

Qualitative research methods were criticized as subjective, biased, non relevant and unscientific as a research tool in the past, especially in natural sciences and did not have a place in medicine. To overcome these criticisms more vigorous methods were applied in all aspects of qualitative research. But now qualitative research is recognized as an important research method and increasingly used in Medicine, especially in Public Health (55, 56, 60).

Kenneth Howe (1988) argues that in any research only a few things are based on scientific grounds and the bulk of the study is based on common sense and previous experiences of logic inherent in the problem definition. According to him even the scientific method or positivist approach is also heavily dependent on subjective influences of the researcher and affects the topic under study and its outcome (57). Malterud (2001) adds to this stating

“medical discipline is founded on scientific knowledge, but clinical decisions and methods of patient care are based on much more than results of the controlled experiments” (56). In medicine the diagnosis or the understanding of the disease is more dependent on quantitative methods but on the other hand understanding the patient is an entirely different matter.

Patients have feelings and opinions and they differ in attitude, behavior and character.

Qualitative studies are essential to understand these phenomena among the patients, doctors and other health care workers (56).

The steps in a qualitative research design are interdependent and overlapping especially because the research design would change and reshape as the information is gathered (60).

(25)

Several factors influence the design of a research project; theoretical and conceptual frame work, purpose or the objectives of the study, which method of data collection is used?, who are the participants?, ethical standards, and finally the method of analysis and dissemination of the information collected (60).

According to Rubin (2005) in the interpretivist paradigm research questions were not decided at the beginning but they emerge during the process of the interview and should be pursued further (67). So the position is that the researcher must have a high level of tolerance towards any uncertainties that could occur especially at the beginning of the research and then with time the researcher would be able to reshape his or her previous ideas and come up with different research questions later on with the progress of the research project (67).

Three main data collection techniques are used in qualitative studies especially in interpretive constructionist paradigm; participant observation, focus group discussions and in-depth interviews (60). In ‘participant observation’ method the researcher would observe the behavior of the participants and the researcher has the choice to sit separate from the participants or to become part of the group and mingle with the study participants. The main focus is to observe the behavior and interactions of the participants in their natural setting such as a work place or home.

A ‘focus group’ is a group of participants preferably six to eight people chosen by the researcher who would sit with the researcher and discuss a common theme of interest (58, 60).

The researcher would not be actively participating in the discussion, rather the researcher would be the moderator and direct and encourage the free flow of ideas from all the participants. In contrast to these two methods in an in-depth interview the researcher would interview individual participants with a face to face interview. Depending on the research theme, the availability of time and resources, the interview could take few minutes to several hours and sometimes the researcher would meet the participant over and over again during the course of the research project, until the researcher is contented about all the possible information that could be obtained from the participant (60, 67).

The main objective of interviewing is to gain the interviewees’ interpretation of the experiences and perceptions of the research topic under investigation. The personal characteristics of the interviewer as well as the interviewee influence the outcome of an

(26)

interview and interviewer also contributes actively to the interpretation of the interview.

Because the interviewer could influence the interviewee in an interview it is important that the interviewer poses his or her questions in a manner that the interviewee could answer without limitations on his or her feelings and opinions (67).

There is a great debate on bias in qualitative methods and bias in research is a concern for any researcher. Especially in qualitative research the outcome of the research is influenced by the bias of the researcher depending on the researcher’s cultural, ethnic, political and religious views. To reduce the bias of the project researchers suggest to increase the rigour in qualitative methods. The use of a tighter research design is called rigour. Several ways of increasing the rigour are suggested; using triangulation methods, using a valid theoretical basis, being reflexive about the biases and perspectives of the researcher and acknowledgment of contradictory interpretations (54). In addition when designing a research reflection on several key issues are also important to reduce bias. Objectivity (influence of the researcher and the participant on the outcome of the research), validity (authenticity of the data collected), reliability (acceptance of the outcome of the study), subjectivity (perspective of the researcher) and sampling are very important (54, 67).

Reliability and validity of the information gathered play an important part in acceptability of the findings of a research. Several ways are suggested to increase the reliability of the data gathered for a qualitative study. Training the researcher on interview techniques prior to the start of the project, using standardized questions and applying transcription rules when transcribing the data are some of the methods suggested. Proper documentation is also important in increasing the reliability of the data gathered (58).

The validity of the information gathered is also important when disseminating the findings.

The researcher must make sure that the interpretation of the data shows the true ideas and opinions of the participants. According to some of the literature, if the participant’s presentation is a narrative then it is regarded as valid (58). But on the other hand it is argued that this is applicable for only a very limited number of situations. Some authors suggest involving the participants to validate the findings. The researcher would meet the participant for a second time to get the response of the participant on the analyzed data. Another way is to compare the findings with the available literature and find some common grounds in theory

(27)

and findings. Finally triangulation can also be used for validation of the information gathered (58).

Generalization of the findings is another hotly debated topic in social sciences as well as natural sciences. Due to the subjective nature of the methodology qualitative research analysis is a challenging task. This is further intensified by the fact that the researcher and the interviewee both could influence the outcome of the research with their biased subjective views. On the other hand that is what the researcher is interested in when conducting a qualitative study. But without generalization the applicability of the findings of a study would become ineffective. Therefore the selection of a representative sample from the population under study would be essential for generalization of the findings (58). On the other hand in qualitative research the depth and insight of a topic is more important than the breadth, so generalization becomes more of a conceptual concept than a numerical one (60).

Selection of a representative sample from the population would not necessarily be an objective in a qualitative study and given the nature and the size of the sample it would not be practical either. The researcher decides which person to be included in the study or not.

Therefore the very nature of qualitative research design is not about generalization of the findings to the wider population but more of a deeper and richer understanding of the topic under study (58, 60).

Selection of subjects for a research project or ‘sampling’ is a challenge in any research.

Unlike in quantitative methods where strict selection criteria can be used, in qualitative research the selection criteria are comparatively more flexible. In qualitative methods selection of participants can change with the progress of the research project. Usually the selection of participants would be carried out until the researcher is convinced that all aspects of the research topic were covered. The researcher has the flexibility to select participants who would provide the richest information for the research. Several main sampling methods are used in qualitative methods. For example, in ‘extreme sampling’ outstanding cases will be selected as participants, in ‘homogenous sampling’ method a subgroup of participants who has the same type of experiences will be selected such as a group of nurses or doctors, in

‘convenient sampling’ the most available and easiest to reach subjects are selected as study participants (60).

(28)

Qualitative study among Sri Lankan immigrants living in Oslo

In my qualitative study also I used several strategies to increase the reliability and validity of the data. Being a novice for the qualitative research and trained as a quantitative person it was a very challenging task for me to look at things with a qualitative perspective, at least at the beginning. I had several classes on qualitative methods. I used a semi structured interview guide for my research and I did two pilot interviews with two Sri Lankan immigrants before I started the project. I documented all the data meticulously as possible. I also used theoretical background for my project. But I did not meet the interviewees for a second time.

Out of the several theoretical approaches described above in my study I inclined more towards the interpretive constructionist theory and focused the research on the experiences of the participants’ interaction with their family, friends and social environment.

My qualitative study included Sri Lankan born first generation immigrants who are currently living in Oslo. The selection criteria for our study had several components.

 Sri Lankan born immigrants living in Oslo

 Age preferably between 31 to 60 years

 Participant is living in Norway at least more than three years.

These criteria were selected because we wanted to make the qualitative study as a complementary study for the larger quantitative survey and I wanted to select a sample of participants as similar as possible to the quantitative study. All the subjects who participated in the quantitative survey were living in Oslo and the age limits were between 31 and 60 years. I used data from participants who had lived at least 3 years in Norway, this is because we were calculating SAD prevalence and it is an objective that the participants had lived minimum of three years in a given location to calculate SAD prevalence.

I used purposive and snow ball sampling method to recruit my participants. In purposive sampling method in contrast to convenient sampling method the researcher selects subjects who can provide the richest information for the research topic (54, 60). Snow ball sampling is a commonly used method to reach and find subjects who are not known to the researcher.

Usually word of mouth and social networks are used to recruit participants in this manner and it has been proven effective in reaching the participants (54, 60). Being a Sri Lankan myself I

(29)

personally know many Sri Lankan immigrants living in Oslo and I have a fairly large network of friends among Sri Lankan immigrants. At the beginning of the research I talked to many of my friends and let them know what I was doing. I prepared an information leaflet in Tamil and English, inviting participants to take part in the project which was approved ethically by the REK (see annex 2, 3). Copies of this leaflet were distributed among many of my friends and acquaintances at the start of the project. Many potential subjects were introduced to me by my friends and finally my study included four males and four females.

The first participant in the study introduced me to several other participants and I interviewed one of these participants as well. The fifth participant was introduced to me by the fourth participant. So four of the participants were recruited by snow ball sampling method and the rest of the participants were introduced to me by some of my friends and acquaintances. I continued with my interviews with recruitment of participants one by one and I stopped recruiting any more participants when I felt that I was not getting any more new information from recruiting anymore participants (data saturation).

The study included semi structured in-depth interviews with all the eight participants. The interviews lasted on average about 60 to 90 minutes. Interviews were done at places that were convenient for each individual participant. Some interviews were carried out at their homes and some were done in cafes or work places. Except for two interviews all the other six interviews were carried out in the evening after work because the participants had no other free time. The interviews were carried out during the winter months of 2011 from October to December.

3.2.1 Ethical considerations

The qualitative study was approved by the Regional Committee for Medicine and Health Research Ethics, Oslo (REK).

Ethical consideration is one of the most important factors in a research study. The Sri Lankan immigrants represent a minority ethnic group in Norway and my qualitative study focused on mental health issues, which can be considered as very sensitive topics in a research. The ultimate choice of study design depends on ethical considerations, in addition to, aims and availability of resources (61). The safety of the subjects’ life, privacy and integrity are some of the most essential issues in relation to ethical considerations.

(30)

In any research that involves human subjects it is especially important to ensure that the participant has the right to withdraw from the research at any time without incurring any repercussions to him or her. Usually in quantitative surveys a written informed signed voluntary consent is obtained prior to the beginning of the research.

In qualitative studies obtaining a written consent signed by the participant poses several challenges to the researcher. Unlike quantitative studies qualitative research involves only few participants. So securing the anonymity and confidentiality of the subjects is even more challenging in a qualitative study. The researcher assures the participants full anonymity and confidentiality before the research begins. Most of the time name, sex and other personal information of the participants are not collected and even the location or the interview set up is changed when the results are published so as not to identify the participants. So obtaining the signature on a legal looking form could jeopardize the confidentiality that the researcher builds up with his or her participants (67).

The researcher must all the time safeguard the promise of confidentiality and anonymity of the participant of the study. Once the results are published the information should not be traced back to the subjects. The confidential information of the subjects should not be divulged to anyone outside the project group. At the same time the researcher should be able to publish the results in a way that the data are valid. This is especially important if the topic under investigation is sensitive in nature (54, 67, 68).

There are many thousands of Sri Lankan immigrants living in Oslo. But the community is so much interconnected it will be easy to identify a person even with a little description. So I had to be extra careful when publishing my data and results to safe guard the anonymity of the participants in my study.

I gave the participants a full and complete description regarding the study and they could opt to withdraw from their participation at any time from the study (See annex 2 and 3). I assured them that the confidentiality and anonymity of them will be safeguarded and none of their personal details will be published or stored with the information gathered. I also will destroy all the data files once the results are published. I did not collect their personal information such as name and address which can be referred back to them. I have also not mentioned the exact location of the interviews and certainly will be discrete when I publish their individual characteristics such as gender when I quote them in my published articles.

(31)

The participants were not paid any compensation for their contribution to the study.

Reimbursement for the time participants spent for a study often arises in Focus Group Discussions (FGD) (60) and it is better to follow the local customs in such an instant. On the other hand, paying for participation in a study would result in recruiting participants with a bias view (60). The customs and traditional values of Sri Lankans is that they usually do not expect money or gifts for a help that they rendered for someone and it would be even considered as an insult. Thus I did not pay any money or gave any gifts to the participants.

(32)

4 Reflexivity

Reflexivity is the critical self awareness of the researcher during a research study (54, 59, 60).

The researcher sees the research project thorough his perspective and this influences the research from the start to the end of the project. The researcher is the one who decides on the research topic, which methods to use, how to recruit subjects and gather data, analyze, interpret and finally disseminate the information gathered. So the outcome of a research is influenced by the researcher from the inception of the project (54, 60).

In qualitative studies the researcher himself is a valuable asset with the knowledge, skills and the experience he or she brings to the field. But at the same time he or she should have an awareness of the influence he could exert on the study (68). The researchers position in the research in regard to his or her education, experiences, culture and background, personal attitudes and characteristics and pre-understanding influences the way he conducts the interviews and direct the participants towards a given topic of discussion (54).

In the past when the researcher’s influence was regarded as an unwanted bias, the researcher did everything possible to minimize this effect when doing research, but now the trend is to acknowledge this fact and in qualitative studies the researcher is considered as an active participant in the process of information gathering and the researcher’s skills and experience as a communication partner is regarded as an asset for the project. However, when sensitive topics are brought in to the discussion it is acknowledged that if both the researcher and the participant are of the same gender it would be more advantageous (60, 67).

I acknowledge the fact that my pre-understanding and prejudices as a medical doctor and a public health researcher on the one hand and as a Sri Lankan on the other hand would affect the way I design, conduct and analyze the research project and final interpretation of the results. For example I have my own opinions and feelings of how mental health is affected by certain factors. So avoiding, disputing the participants of their feelings and experiences when talking about mental health was a real challenge for me. Also as a fellow Sri Lankan my political, social and cultural opinions were not the same as some of the participants and when social and cultural topics were discussed or analyzed the influence I have as a researcher could not be denied. On the other hand pre-understanding and prejudices are issues that have shaped who a person is and denying the presence of these could jeopardize the outcome of a research.

Seasonal Affective Disorder and associated Factors among Sri Lankan Immigrants living in Oslo